The wear on the drive splines of the No. 1 element of the pump resulted from the mismatching of chromed and non-chromed components and the progressive breakdown of a nitride white layer. Overhaul procedures amended in 1983 required replacement of the non-chromed components, but the pump had not attained TBO time limits during its 20-year in-service life, and no directives had been issued to require pre-overhaul replacement. Functional testing and repair of the pump, conducted in 1992, was limited in scope and did not require complete disassembly and re-build. Consequently, the progressive wear went undetected. When the No. 1 element drive splines disengaged and the No. 1 element check valve jammed open, pump flow from the No. 2 element re-circulated within the pump and little or no fuel flow was provided to the engine. As failed spline pieces jammed between spline remnants in the No. 1 element of the pump, it is likely that momentary engine power reductions would be followed by short periods of normal power. Eventually, the wear would have progressed to a point that temporary engagement of the drive was impossible and the engine lost all power. The pilot was faced with power interruptions and engine indications that were initially difficult to analyse. The helicopter was being operated without a fuel pressure differential switch, and the attendant bypass warning system was therefore inoperative. However, because post-accident examination of the fuel filter showed no evidence of filter contamination or restriction, if the pressure differential switch had been installed and the system had been working, the fuel filter bypass warning would not have been activated. The absence of the pressure differential switch and warning system would not have contributed to the occurrence. While there was insufficient information to determine the exact weight and balance of the helicopter, the maximum upper weight limit of 2 530 pounds that was approximated using the POH best range charts is less than the certificated maximum gross weight of 2 550 pounds. Because it was unlikely that the maximum range fuel flows could have been achieved during the operations flown that morning, more fuel would have been used, there would have been less fuel on board and the weight of the helicopter would likely have been somewhat less than 2 531.5 pounds. The helicopter was being flown at less than the maximum gross weight, within its flight envelope, and an autorotation should have been possible. However, the helicopter would have been significantly heavier than it was at the time that the pilot had practised autorotations in training. Company policy required that the pilot transit at an altitude of 500 feet agl, and although there is no hard information to confirm that the pilot climbed to the transit altitude, there is no information that he did otherwise on this flight. It is likely that he was at the transit altitude when the engine problems began. Since he was flying downwind when the engine-driven fuel pump failure occurred, the correct procedure would have been to turn into wind. There was no information available to explain why the pilot did not turn immediately to prepare for an emergency landing into wind. However, it is possible that the time that he used to reassure the passengers and time that he may have spent analysing the power interruptions resulted in an altitude or airspeed loss that eliminated the possibility of a turn into wind. Therefore, when the engine failed completely, the pilot was faced with executing an autorotation with a strong tailwind, over a relatively featureless surface. His perception of forward speed and cues to judge height above the surface would have been significantly different from his normal experience in practice autorotations. There was little information as to the manner in which the pilot reacted to these abnormal conditions and flew the autorotation; however, the lack of main rotor system rotational energy at impact indicates that the pilot did not maintain rotor rpm throughout the manoeuvre. The available information indicates that the helicopter appeared to be flared, tail down, at about tree-top height before its abrupt descent. Once this flight attitude was reached without engine power and at low rotor rpm, there was insufficient airflow through the rotor to sustain rotor rpm. The rotor rpm would have decayed further, and without rotor rpm, the pilot was unable to control the helicopter. The helicopter tilted to its left side and descended abruptly as it was carried forward by the residual speed and tailwind. The following Engineering Branch reports were completed: LP 075/98 - Analysis of Plastic and Asbestos LP 110/98 - Engine Driven Fuel PumpAnalysis The wear on the drive splines of the No. 1 element of the pump resulted from the mismatching of chromed and non-chromed components and the progressive breakdown of a nitride white layer. Overhaul procedures amended in 1983 required replacement of the non-chromed components, but the pump had not attained TBO time limits during its 20-year in-service life, and no directives had been issued to require pre-overhaul replacement. Functional testing and repair of the pump, conducted in 1992, was limited in scope and did not require complete disassembly and re-build. Consequently, the progressive wear went undetected. When the No. 1 element drive splines disengaged and the No. 1 element check valve jammed open, pump flow from the No. 2 element re-circulated within the pump and little or no fuel flow was provided to the engine. As failed spline pieces jammed between spline remnants in the No. 1 element of the pump, it is likely that momentary engine power reductions would be followed by short periods of normal power. Eventually, the wear would have progressed to a point that temporary engagement of the drive was impossible and the engine lost all power. The pilot was faced with power interruptions and engine indications that were initially difficult to analyse. The helicopter was being operated without a fuel pressure differential switch, and the attendant bypass warning system was therefore inoperative. However, because post-accident examination of the fuel filter showed no evidence of filter contamination or restriction, if the pressure differential switch had been installed and the system had been working, the fuel filter bypass warning would not have been activated. The absence of the pressure differential switch and warning system would not have contributed to the occurrence. While there was insufficient information to determine the exact weight and balance of the helicopter, the maximum upper weight limit of 2 530 pounds that was approximated using the POH best range charts is less than the certificated maximum gross weight of 2 550 pounds. Because it was unlikely that the maximum range fuel flows could have been achieved during the operations flown that morning, more fuel would have been used, there would have been less fuel on board and the weight of the helicopter would likely have been somewhat less than 2 531.5 pounds. The helicopter was being flown at less than the maximum gross weight, within its flight envelope, and an autorotation should have been possible. However, the helicopter would have been significantly heavier than it was at the time that the pilot had practised autorotations in training. Company policy required that the pilot transit at an altitude of 500 feet agl, and although there is no hard information to confirm that the pilot climbed to the transit altitude, there is no information that he did otherwise on this flight. It is likely that he was at the transit altitude when the engine problems began. Since he was flying downwind when the engine-driven fuel pump failure occurred, the correct procedure would have been to turn into wind. There was no information available to explain why the pilot did not turn immediately to prepare for an emergency landing into wind. However, it is possible that the time that he used to reassure the passengers and time that he may have spent analysing the power interruptions resulted in an altitude or airspeed loss that eliminated the possibility of a turn into wind. Therefore, when the engine failed completely, the pilot was faced with executing an autorotation with a strong tailwind, over a relatively featureless surface. His perception of forward speed and cues to judge height above the surface would have been significantly different from his normal experience in practice autorotations. There was little information as to the manner in which the pilot reacted to these abnormal conditions and flew the autorotation; however, the lack of main rotor system rotational energy at impact indicates that the pilot did not maintain rotor rpm throughout the manoeuvre. The available information indicates that the helicopter appeared to be flared, tail down, at about tree-top height before its abrupt descent. Once this flight attitude was reached without engine power and at low rotor rpm, there was insufficient airflow through the rotor to sustain rotor rpm. The rotor rpm would have decayed further, and without rotor rpm, the pilot was unable to control the helicopter. The helicopter tilted to its left side and descended abruptly as it was carried forward by the residual speed and tailwind. The following Engineering Branch reports were completed: LP 075/98 - Analysis of Plastic and Asbestos LP 110/98 - Engine Driven Fuel Pump The company records indicated that the helicopter was inspected, certified and maintained in accordance with approved schedules. The helicopter was being operated without a fuel filter pressure differential switch or fuel filter bypass warning system; however, post-accident inspection confirmed that the fuel filter was free from contamination and the non-functioning filter bypass warning system would not have contributed to the occurrence. The damage sustained by the main rotor and tail rotor components and their respective control systems resulted from the impact of the helicopter and its components with the ice. The engine-driven fuel pump failed as a result of a progressive failure of the drive splines of the No. 1 element of the pump and subsequent metal contamination of the No. 1 element check valve. The failure of the pump splines resulted from the combined effects of continued operation of chromed drive shaft splines against non-chromed spur gear splines, and the effects of the breakdown of a nitride 'white layer'. The pairing of a chromed drive shaft against non-chromed spur gear splines resulted from the incorporation of Allison Engine 250-C18 CEB-161, which likely occurred when the pump was overhauled some time before 15 November 1976. In 1983, the pump manufacturer amended the overhaul manual to reflect deletion of pump models 024731-112 and 024731-113 which contained the P/N 02-14624 (non-chromed) drive spur gears. Current pump models 024731-132, -133, -135, and -136 require installation of P/N 02-16057 (chromed) drive spur gears. The pump does not have a calendar life overhaul limit, and there were no directives issued to require change out of the spur drive gears before TBO time limits. Despite being in service for more than 20 years, the pump had not accumulated sufficient service time to require an overhaul. The progressive failure of the pump likely resulted in short-duration power deviations that became more exaggerated until the engine suffered a complete loss of power. The weight and center of gravity of the helicopter were within the prescribed limits. The pilot was qualified and certified for the flight in accordance with existing regulations. There was no indication that the pilot's performance was degraded by physiological factors. 1It could not be determined why the pilot did not maintain sufficient rotor rpm to control the helicopter while conducting an autorotation.Findings The company records indicated that the helicopter was inspected, certified and maintained in accordance with approved schedules. The helicopter was being operated without a fuel filter pressure differential switch or fuel filter bypass warning system; however, post-accident inspection confirmed that the fuel filter was free from contamination and the non-functioning filter bypass warning system would not have contributed to the occurrence. The damage sustained by the main rotor and tail rotor components and their respective control systems resulted from the impact of the helicopter and its components with the ice. The engine-driven fuel pump failed as a result of a progressive failure of the drive splines of the No. 1 element of the pump and subsequent metal contamination of the No. 1 element check valve. The failure of the pump splines resulted from the combined effects of continued operation of chromed drive shaft splines against non-chromed spur gear splines, and the effects of the breakdown of a nitride 'white layer'. The pairing of a chromed drive shaft against non-chromed spur gear splines resulted from the incorporation of Allison Engine 250-C18 CEB-161, which likely occurred when the pump was overhauled some time before 15 November 1976. In 1983, the pump manufacturer amended the overhaul manual to reflect deletion of pump models 024731-112 and 024731-113 which contained the P/N 02-14624 (non-chromed) drive spur gears. Current pump models 024731-132, -133, -135, and -136 require installation of P/N 02-16057 (chromed) drive spur gears. The pump does not have a calendar life overhaul limit, and there were no directives issued to require change out of the spur drive gears before TBO time limits. Despite being in service for more than 20 years, the pump had not accumulated sufficient service time to require an overhaul. The progressive failure of the pump likely resulted in short-duration power deviations that became more exaggerated until the engine suffered a complete loss of power. The weight and center of gravity of the helicopter were within the prescribed limits. The pilot was qualified and certified for the flight in accordance with existing regulations. There was no indication that the pilot's performance was degraded by physiological factors. 1It could not be determined why the pilot did not maintain sufficient rotor rpm to control the helicopter while conducting an autorotation. The helicopter engine suffered power excursions and finally a complete power loss as a result of the failure of the engine-driven fuel pump. The helicopter crashed onto the frozen lake surface when the pilot allowed the rotor rpm to decay and lost control of the helicopter while attempting an autorotation.Causes and Contributing Factors The helicopter engine suffered power excursions and finally a complete power loss as a result of the failure of the engine-driven fuel pump. The helicopter crashed onto the frozen lake surface when the pilot allowed the rotor rpm to decay and lost control of the helicopter while attempting an autorotation. The company has indicated that it will no longer be using the dual element fuel pumps on its helicopters.Safety Action The company has indicated that it will no longer be using the dual element fuel pumps on its helicopters.